JPH0895106A - Equipment and method for optical signal transformation - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a device and a method for correcting dispersion of a chromatic dispersion, which is led by converting the optical signal. SOLUTION: This device has a non-linear converting medium (optical phase conjugator 20) arranged inside an optical signal passage of the optical signal. The converting medium receives the optical signal, and generates the converted optical signal. At least one dispersion correcting unit 22 is arranged in the signal passage so as to supply the appropriate chromatic dispersion variable for off-set of a part of the chromatic dispersion, which is led to the converting signal by the non-linear converting medium. The non-linear converting medium is a dispersion deviating fiber at an appropriate length to be used for phase conjugation of the input optical signal inside an optical system, which is performed by using an optical phase conjugation for eliminating the non-linear characteristic of the fiber, and/or frequency deviation of the input optical signal. The dispersion correcting unit 22 is arranged inside the optical signal passage in front or the rear of the non-linear converting medium, or dispersed in the medium.

Description

Detailed Description of the Invention

[0001]

FIELD OF THE INVENTION The present invention relates to improvements in optical communication systems. More specifically, the present invention relates to chromatic dispersion correction in optical communication systems that use optical phase conjugation or other types of optical signal conversion.

[0002]

2. Description of the Related Art In an optical communication system using an optical fiber as a transmission medium, chromatic dispersion and fiber non-linearity are major obstacles for achieving a high system data transmission rate and a long transmission distance without a repeater. Chromatic dispersion (often referred to simply as "dispersion") is called the phenomenon where the speed of an optical signal over an optical transmission medium such as fiber changes as a function of the optical signal wavelength.

The problem of chromatic dispersion is particularly important in standard single mode fiber (SMF), which forms the majority of fiber optic infrastructures present in the world. Standard SMFs generally have a well-defined dispersion for wavelengths longer than zero dispersion, but exhibit zero dispersion at wavelengths around 1300 nm.

Dispersion can be expressed in the form of variations in the propagation constant of a fiber with frequency. The first-order and second-order group velocity dispersions are related to the second and third derivatives of the fiber propagation constant with respect to the angular frequency ω or β ₂ and β ₃ , respectively. Dispersions higher than this can be approximated to zero for most applications. When used in the context of lightwave transmission systems, first and second order dispersions are usually expressed in the form of dispersion over wavelength.

Therefore, first order group velocity dispersion is commonly referred to as the change in pulse transit time in a unit length of fiber for changes in pulse wavelength. in this case,
The symbol D (λ) is often used to indicate first order group velocity dispersion. This unit is typically picoseconds / nanometer kilometer (ps / nmkm). The secondary group velocity dispersion is D (λ) using the unit of ps / nm ² · km.
Is shown as the derivative with respect to the wavelength of.

An important fiber nonlinearity that limits transmission capability is the Kerr effect. In this case, the refractive index increases with the intensity of the input optical signal. The change in the index of refraction of the fiber modulates the phase of the optical signal passing through the fiber, which
Redistribute the signal frequency spectrum. In systems where an optical signal modulates itself, this phenomenon is commonly known as self-phase modulation. Self-phase modulation produces a low frequency towards the leading edge of the optical signal pulse and a high frequency towards the trailing edge.

In multi-channel systems where one signal causes modulation of another, this phenomenon manifests itself as either cross-phase modulation or four-photon mixing. In both single-channel and multi-channel systems, the resulting changes in frequency distribution are translated into amplitude modulation by fiber dispersion. Thus, the interaction between chromatic dispersion and non-linearities such as the Kerr effect results in increased optical signal distortion as a function of transmission distance. Therefore, for long distance communication over optical fiber, dispersion and non-linearity must be controlled, corrected or suppressed.

Midspan optical phase conjugation is a technique used to mitigate chromatic dispersion effects in optical systems. In fact, since the phase conjugation of an optical pulse is the time reversal of the pulse, an optical phase conjugator placed at the center of the fiber span will give the first-order group velocity dispersion of the first half of the span and the conjugate signal It can be corrected by the same first order group velocity dispersion produced as it propagates along the half.

For example, A. Gnauck, R. Jopson and R. De
rosier, "10 Gbit / s 360km Transmission over Dispers
ive Fiber Using Midsystem Spectral Inversion, "IEE
E Photonics Technology Letters, Vol. 5, No. 6, pp.
663-666, June 1993; and S. Watanabe et. Al., "Com
pensation of Chromatic Dispersion in a Single-mode
Fiber by Optical Phase Conjugation, "IEEE Photoni
cs Technology Letters, Vol. 5, No. 1, pp. 92-95, J
See anuary 1993.

Dispersion correction using optical phase conjugation has also been demonstrated for wavelength division multiplexed (WDM) systems. For example, A. Gnauck, R. Jopson, P. Iannone and R. Dero
sier, "Transmmision of two wavelength -multiplexed
10 Gbit / s channels over 560km of dispersive fibe
r, "Electronics Letters, Vol. 30, No. 9, pp. 727-7
See 28, April 1994.

[0011] US patent application Ser. No. 08/120014 discloses that the effects of fiber span nonlinearity can be corrected using optical phase conjugation. One example of an optical system described in this specification adjusts the optical signal output in a fiber span by selecting a suitable number, spacing and output of in-line amplifiers so that the effects of fiber nonlinearity are corrected. To do.

For a more detailed description of fiber nonlinearity correction using optical phase conjugation, see, for example, C. Kurtzke and
A. Gnauck, "How to Increase Capacity beyond 200Tbi
t / s.km without Solitons, "ECOC '94 Proceedings, Vo
l. 3, Postdeadline Paper No. ThC 12.12, pp. 45-48,
Montreux, Switzerland, September 1993; W. Pieper
et al., "Nonlinearity-inspective standard-fiber t
ransmission based on optical-phase conjugation in a
semiconductor-laser amplifier, "Electronics Lette
rs, Vol. 30, No. 9, pp. 724-726, 1994; and S. Wat
anabe and T. Chikama, "Cancellation of four-wave m
ixing in multichannel fiber transmission by midway
optical phase conjugation, "Electronic Letters,
Vol. 30, No. 14, pp. 1156-1157, July 1994.

Another concern in the optical system disclosed in US patent application Ser. No. 08/120014 is the chromatic dispersion effect introduced by the optical phase conjugator itself. For example, a common method of phase conjugating optical signals is 4-photon mixing in the length of dispersion-shifting fiber (DSF). DS
Efficient mixing at F generally requires proper phase matching. Phase matching is done by placing the pump signal wavelength near the zero dispersion of the DSF, or, if two pump signals are used, placing the average of these wavelengths near zero dispersion. .

However, under these conditions, it has been discovered that chromatic dispersion in the DSF phase conjugator significantly distorts the phase conjugate output signal. Zero dispersion at 1550 nm, pump signal at wavelength 1550 nm and wavelength 1554 nm
In a phase conjugator consisting of a 20 km DSF with an input signal at, the phase conjugate output signal is generated at a wavelength of approximately 1546 nm. Typical second-order dispersion value of DSF is 0.0
When it is 8 ps / nm ² · km, the DSF is (1546
(nm-1550 nm) × 0.08 ps / nm · km or −0.32 ps / nm · km at the wavelength of the conjugate signal, producing a total first-order dispersion.

Therefore, the conjugate signal at the output of the DSF exhibits a substantial additional chromatic dispersion of (-0.32 ps / nm * km) * (20 km) or -6.4 ps / nm. Therefore, the chromatic dispersion generated in the optical phase conjugator is so great that it reduces or eliminates the benefits gained from phase conjugation. Similar concerns include DSF, single mode fiber (S
It also applies to other types of optical signal converters using MF) or other alternative non-linear conversion media.

[0016]

SUMMARY OF THE INVENTION It is therefore an object of the present invention to provide an optical phase conjugator, frequency shifter or other signal converter that includes correction for the chromatic dispersion introduced as a result of the conversion.

[0017]

SUMMARY OF THE INVENTION The present invention is an optical phase conjugate,
An apparatus and method for correcting chromatic dispersion introduced in a non-linear medium that performs frequency shift and / or other signal conversion processing. An example of the device according to the invention comprises a non-linear conversion medium and at least one dispersion compensator. The non-linear conversion medium is arranged in the optical signal path, can receive an optical signal, and can generate a converted optical signal from this signal.

The dispersion corrector is arranged in the optical communication line and provides an amount of chromatic dispersion suitable for shifting a part of the chromatic dispersion introduced into the converted signal by the nonlinear conversion medium. The non-linear conversion medium is, for example, an optical fiber of a predetermined length used to effect a frequency shift of the input optical signal via optical phase conjugation and / or four-photon mixing. Alternatively, the non-linear medium may be an active or passive semiconductor, a non-linear crystal or other non-linear medium suitable for converting an optical signal into frequency and / or phase. The dispersion corrector can be placed either before or after the nonlinear medium in the optical signal path.
In another embodiment of the invention, the dispersion corrector can also be distributed in the non-linear conversion medium.

An example of the method according to the present invention is the step of inputting an optical signal into a nonlinear conversion medium arranged in the signal path of the optical signal, the step of generating the converted optical signal from the optical signal in the nonlinear medium, Correcting a portion of the chromatic dispersion introduced by the medium into the converted signal by providing an offset amount of chromatic dispersion in the optical signal path. The step of correcting the chromatic dispersion introduced by the non-linear medium comprises providing additional chromatic dispersion either before or after being dispersed in the non-linear medium. The nonlinear medium is
For example, it could be part of an optical phase conjugator used to correct the interaction between nonlinearity and chromatic dispersion in the optical fiber span.

The present invention corrects chromatic dispersion introduced in signal conversion processes such as optical phase conjugation and / or frequency shift of optical signals. Compensating for this additional chromatic dispersion is best obtained in applications using signal transformations such as midspan optical phase conjugation. Proper compensation of the dispersion introduced by the non-linear medium in systems that use optical phase conjugation to compensate for fiber non-linearities maximizes fiber span bit rate range products. The correction can be provided easily and inexpensively in a variety of different optical systems using any of a number of nonlinear conversion media such as dispersion-shifted fiber (DSF), semiconductor lasers, semiconductor laser amplifiers and nonlinear crystals. You can

[0021]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The optical phase conjugation of the present invention will be described with reference to FIG. FIG. 1 illustrates an optical communication system 1 incorporating optical phase conjugation to correct fiber dispersion and / or non-linearity with additional dispersion correction according to the present invention.
It is a schematic block diagram which shows an example of 0. The system 10
An optical signal transmitter 12 is provided at one end of an optical fiber span composed of a large number of fiber amplifiers 14 and long optical fibers 16.

The amplifier 14 is, for example, an erbium-doped fiber amplifier (EDFA). Erbium-doped fiber amplifiers are spaced to correct for optical fiber attenuation and to approach lossless power distribution across the fiber span. Another fiber span suitable for use in the present invention need not have such a distribution, and further need not have an in-line amplifier. For example, transmitter 12
However, amplifier 14 can be eliminated if it provides sufficient optical signal power to communicate efficiently over the span.

The optical signal receiver 18 is located on the opposite side of the optical fiber span. Therefore, the fiber span provides an optical signal path between transmitter 12 and receiver 18. As used herein, the term "optical signal path" includes any optical transmission medium through which an optical signal passes, such as fiber spans, optical waveguides or free space. The system 10 also has an optical phase conjugator 20. Optical phase conjugator 20 produces a phase conjugation of the input signal to correct for chromatic dispersion effects within the fiber span.

A more detailed description of the use of optical phase conjugation to correct for dispersion can be found in, for example, A. Yariv et al., "
Compensation for Channel Dispersion by Nonlinear O
ptical Phase Conjugation, "Optics Letters, Vol. 4,
No. 2, pp. 52-54, February 1979.

System 10 is described in the above-referenced US patent application Ser.
A phase conjugator 20 is also used to correct the fiber nonlinearity in the manner described in 8/120014. The fiber optic span of FIG. 1 can be considered to consist of a single segment having a first portion and a second portion. A single optical phase conjugator 20 is located between the first and second parts of the span. The first and second parts of the span need not be exactly the same length.

The term "segment" used above to describe an optical fiber span relates to the number of optical phase conjugators in the span. In the embodiment shown in FIG. 1, the fiber span has only a single optical phase conjugator 20 located between the first and second parts of the span. Therefore, for a single phase conjugator, the span and segment are one and the same. However, in other cases it is desirable to divide the span into several segments. Each segment has a phase conjugator between these first and second parts.

In systems which use the phase conjugator 20 to correct only first-order group velocity dispersion, the phase conjugator 20 is typically near the center of the system or at L / 2 points of a fiber span of length L. It will be located nearby. Such a system is called a linear system and, for convenience, often uses evenly spaced amplifiers 14. In the system described in US patent application Ser. No. 08/120014, the interaction between chromatic dispersion and fiber nonlinearity is corrected by adjusting the power level of the optical signal at various points in the fiber span. It

The adjustment of the optical signal power is performed by adjusting the number of in-line amplifiers 14, the relative position or spacing between the amplifiers 14, and / or
Alternatively, the output power of one or more amplifiers 14 may be selected appropriately. As a result, the conjugator 20 is not located at the exact midpoint of the span. By making power adjustments in this manner, the interaction of chromatic dispersion and fiber nonlinearity in the first portion of a given span segment, for example, provides the same interaction in the second portion of a given span segment. It can be performed by doing.

In these nonlinear systems, in general,
At the output of the phase conjugator 20, the phase relationships between all signals (including all products of nonlinearities) are those phase related conjugations at the input to the conjugator 20. The chromatic dispersion within the phase conjugator 20 undesirably changes these phase relationships.

Phase conjugation of an optical signal is performed by using, for example, four-photon mixing (also called “four-wave mixing”). Four
Photon mixing is a non-linear process in which an input optical signal is mixed with one or more pump signals in a nonlinear conversion medium to produce a mixing product. The non-linear conversion medium comprises, for example, an active element such as a semiconductor laser or a semiconductor laser amplifier, a passive semiconductor or a length of optical fiber. The four-photon mixing process itself is either non-degenerate or degenerate.

In non-degenerate four-photon mixing, two separate pump signals mix with the input optical signal to produce a fourth signal. For an input signal at frequency f _s, a first pump at frequency f _{p1 and} a second pump at frequency f _p2 , this non-degenerate mixing process produces a phase conjugate of the input signal at frequencies f _p1 + f _p2 −f _s . To do. In some forms of degenerate four-photon mixing, the two mixing signals are provided by a single pump.
Thus, for an optical signal at frequency f _s and a pump at frequency f _p , degenerate four-photon mixing produces a phase conjugate of the optical signal at frequency 2f _p −f _s .

Generally, it is important to match the polarization of the input and pump signals before phase conjugation. This is done by using a polarization controller in the signal path of the input and pump signals. Alternatively, the need to control the input and pump signal polarizations can be eliminated by using polarization insensitive optical phase conjugators.
Such a polarization insensitive optical phase conjugator is disclosed in US patent application Ser. Nos. 08/120013 and 08/120011.
No. 8 specification. The present invention uses other non-linear processes, such as three-photon mixing, to perform the signal conversion.

FIG. 1 also has a dispersion corrector 22 located in the optical signal path of the optical signal passing through the fiber span from the optical transmitter 12 to the optical receiver 18. Dispersion corrector 22 provides a certain amount of chromatic dispersion suitable for significantly offsetting the chromatic dispersion introduced by the optical phase conjugator 20 into the conjugate output signal. As mentioned above, a particular type of optical phase conjugator is
A significant amount of chromatic dispersion is introduced in the phase conjugate output signal.
For example, the optical phase conjugator 20 can have a length of dispersion shifted fiber (DSF). This dispersion shifted fiber (DSF) uses a nonlinear four-photon mixing process to produce the phase conjugate of the input optical signal.

When the optical phase conjugator 20 is phase matched,
The chromatic dispersion introduced by the optical phase conjugator 20 is approximately equal to the dispersion that occurs when an input optical signal passes through a nonlinear medium and is phase-conjugated without distortion. This amount of chromatic dispersion is approximately equal to the dispersion introduced into the optical signal passing through the nonlinear medium at the phase conjugate output signal wavelength. In many common optical signal conversion applications, such additional chromatic dispersion significantly degrades the quality of the converted signal output.

The dispersion corrector 22 is designed to provide a sufficient amount of chromatic dispersion to significantly offset the additional chromatic dispersion introduced into the converted signal output by the non-linear conversion medium within the optical phase conjugator 20. ing. The dispersion corrector 22 preferably provides enough dispersion to significantly offset all the dispersion introduced by the phase conjugator 20, but
The dispersion corrector 22 can also supply only the partial offset of the additional dispersion. The dispersion corrector 22 can be placed either before or after the phase conjugator 20 in the system of FIG.

In another embodiment of the present invention, the optical phase conjugator 20 and the dispersion corrector 22 are described in detail below.
Can be combined with a device that provides both phase conjugation and dispersion correction functions. For example, the dispersion corrector 2
The dispersion correction provided by 2 is dispersed in a length DSF at a number of discrete points, or the optical phase conjugator 20
Uses a dispersion-flattened fiber that exhibits minimal dispersion at the desired frequency.

Furthermore, the dispersion compensator 22 can also be designed to provide offset dispersion over a wide range of wavelengths, thereby allowing the phase conjugator 20 to provide the same magnitude of dispersion at a given wavelength. The same dispersion corrector 22 can be used either before or after. Such an embodiment is, for example, a multi-channel WD
It is especially useful in M-light systems.

FIG. 2 is a block diagram of an exemplary optical system 100 according to the present invention that corrects for dispersion introduced by the optical phase conjugation process. The system of FIG. 2 is a three channel WDM system designed to illustrate the improvements in which the dispersion correction technique of the present invention can be used, and limits the present invention to use only in certain optical systems. Do not interpret. The operation of the system of FIG.
The optical signal spectrum of D, the eye diagrams of FIGS. 4A to 4D, and the bit error rate curve of FIG. 5 will be described.

The system 100 has a transmitter 100 that provides three different input optical signals or channels at different but very close wavelengths. The first input optical signal has a wavelength of 155
It is a continuous wave (CW) signal of 5.3 nm. The wavelength of the second input signal (also called the center channel) is 1
555.5 nm and the amplitude is 2.5 of length 2 ²³ -1.
It is modulated with a Gigabit per second non-return to zero (NRZ) pseudorandom bit sequence.

The third input optical signal has a wavelength of 1555.7 nm
Is a CW signal. These three types of input signals are EDFA1
02, and then input to a bandpass filter 104 having a bandwidth of 1 nm to remove unwanted amplified spontaneous emission (ASE) noise. In this particular example, the signal power level is selected to cause non-linear effects in the first portion of the fiber span segment within system 100.

These operations are performed by the above-mentioned US patent application No. 0.
As described in detail in 8/120014,
It can be corrected as the phase conjugate signal propagates in the second portion of the segment. In this embodiment, by properly adjusting the output power of transmitter 101 and / or in-line amplifier 102, the power levels of the first and third signals, as measured at the output of filter 104, are approximately +9 dBm. And the average power level of the modulated second signal is set to about +7 dBm.

Thus, the total signal power of about +13 dBm is supplied to the output terminal of the filter 104. As shown below, the selected power level introduces non-linear distortion in the fiber span. This non-linear distortion is corrected using the optical phase conjugation according to the present invention.

FIGS. 3A and 4A show the baseline optical signal spectrum and eye diagram, respectively, measured at the output of filter 104. The eye diagram in FIG. 4A corresponds to the duplicate bits of the 2 ²³ −1 pseudo-random data stream obtained by demodulating the modulated second optical signal.

Each horizontal axis portion in FIGS. 4A to 4D is 1
Corresponds to a time interval of 00 picoseconds. The modulated second signal is filtered using a fiber Fabry-Perot filter with a bandwidth of 9 GHz to filter the input signal spectrum and a filtered spectrum using an avalanche photodiode (APD) receiver with a bandwidth of 2.5 GHz. Is detected to detect demodulation.

As is clear from FIG. 3A, the optical signal spectrum is relatively clear and exhibits a strong peak at each input signal wavelength. The corresponding eye diagram in FIG. 4A is
It shows a clear separation between high and low logic states (also known as upper and lower rails, respectively) at overlapping bits of the demodulated pseudo-random data stream.

The multi-channel optical signal spectrum shown in FIG. 3A is then converted into a 15.4 km dispersion-shifted fiber (DS) having a zero dispersion wavelength λ ₀ of about 1555.5 nm.
F) into a first length fiber 106. As a result of the selected power level for the input optical signal, the DSF 106 introduces considerable non-linear distortion in the modulated second signal due to four-photon mixing of the three input signals.

FIG. 3B shows three types of input signals and DSF1.
Figure 6 is an optical spectrum measured at the output of DSF 106 showing the additional mixing product resulting from four-photon mixing at 06. FIG. 4B shows an eye diagram generated by demodulating the modulated second signal after being transmitted through the DSF 106. As shown, the upper rail of the eye diagram is significantly distorted. This indicates that the demodulation process cannot fully reproduce the 2 ²³ -1 pseudo-random bit sequence.

Further indicating the degree of distortion introduced into the modulated second signal by the DSF 106 is that using an APD photodiode receiver with a decision threshold fixed at the average signal level, the error rate in the demodulated data stream is erroneous. The Hewlett Packard model number 70842A bit error rate tester used to measure the rate is too large to be synchronized.

Referring again to FIG. The input signal passes through the polarization controller 108 and adjusts the polarization state of the signal.
The signal is then phase-conjugated as follows. About 15
A pump signal having a wavelength λ _{p of} 47.2 nm and a power level of about 0 dBm is generated by the pump signal generator 110, amplified by the EDFA 112 to a power level of about +20 dBm, filtered by the bandpass filter 114, and then ASE. Remove noise.

The polarization of the pump signal is determined by the polarization controller 116.
Is adjusted using. Polarization controllers 108 and 11
6 aligns the input and pump signal polarizations before optical phase conjugation. After that, the pump signal is output from the signal combiner 120.
At the input optical signal. The combined input and pump signals are input to the non-linear conversion medium 122.

In this embodiment, the nonlinear conversion medium 122
Has a zero dispersion wavelength λ ₀ of about 1547.2 nm 2
It consists of a 5.0 km DSF. The non-linear medium 122 produces a phase conjugate of the input signal and a number of other mixing products via the degenerate four-photon mixing and the pump signal. In this embodiment, the phase conjugation is shifted wavelength by wavelength with the phase conjugation center channel shifted to a wavelength of approximately 1539 nm.

The optical signal path after the nonlinear medium 122 is used to separate the desired phase conjugate output signal (including the modulation center channel) from the residual pump signal, the residual input signal and the unwanted mixing product. Has a filter. The band reject filter 124 has a rejection band of 1 nm at the pump wavelength and removes a significant amount (eg, up to 90% or more) of the residual pump signal output at the output of the conversion medium 122.

The bandpass filter 126 has a bandwidth of 4 nm and allows a large number of undesired mixing products, additional residual pump signal output and residual non-conjugate input signal output while passing the phase conjugate input signal. To decay. The optical signal path further comprises a length of single mode fiber (SMF) 1
Also has 30. The SMF 130 provides dispersion correction in the manner described in detail below.

The phase conjugate signal is amplified by the EDFA 134 to a power level substantially the same as the power level (+13 dBm) at the input end of the fiber 106 of the first length. It is then input to the second length fiber 136. Fiber 136 consists of a 14.7 km DSF with a zero dispersion wavelength of about 1540.7 nm. Fiber 136 is selected to compensate for the non-linear distortion introduced by first length fiber 106 in accordance with the techniques disclosed in the aforementioned US patent application Ser. No. 08/120014.

The phase conjugate signal passing through the DSF 136 is input to the bandpass filter 140. In this embodiment, the bandpass filter 140 is approximately 1539 nm.
Is a 9 GHz bandwidth Fiber Fabry-Perot filter designed to pass through the modulation center channel of the H., but not other phase conjugate signals. Receiver 142
Receives a phase conjugate center channel signal, demodulates it and provides a demodulated 2 ²³ -1 pseudo-random bit stream. As mentioned above, a suitable receiver is capable of operating at a bit rate of 2.5 Gbit / sec.
It is a PD receiver.

FIGS. 3C and 4C respectively show the signal spectrum and eye diagram measured with the optical system of FIG. 2 without dispersion correction according to the invention. Therefore, FIG.
C and FIG. 4C correspond to a system without the dispersion corrector 130. As mentioned above, the chromatic dispersion introduced into the nonlinear transform medium of the phase conjugator (eg, DSF 122 in FIG. 2) significantly distorts the output signal.

FIG. 3C shows the phase conjugate signal spectrum measured at the output of DSF 136. As shown, the spectrum at this point contains a number of undesired products, mainly produced by non-linear distortions in the DSF 106, which are left uncorrected by the DSF 136.

FIG. 4C shows the corresponding eye diagram measured by demodulating the center channel at receiver 142. The upper rail of the eye diagram is significantly adversely affected by the non-linear distortion, but compared to FIG. 4B, the non-linear distortion is slightly removed by passing through the DSF 136.

The present invention is based on the following findings.
The hindrance of the DSF 136 to more completely correct the non-linear distortion introduced in the DSF 106 is primarily due to the chromatic dispersion introduced in the phase conjugator (ie, the DSF 122). As mentioned above, this chromatic dispersion can be removed or corrected in a number of different ways. One way is to introduce a corrected chromatic dispersion at the input or output of the phase conjugator.

In the embodiment shown in FIG. 2, an additional amount of positive chromatic dispersion is provided at the output of the conjugator using a dispersion corrector (consisting of 750 m standard SMF). SMF1
30 shows zero dispersion at about 1300 nm, about 1539n
Provides positive dispersion for longer wavelengths, such as m phase conjugate center channel wavelengths. The length of the SMF 130 is 1
The nondispersion introduced by the DSF 122 at 539 nm is chosen to be significantly offset.

3D and 4D respectively show the optical spectrum and eye diagram corresponding to the system of FIG. 2 with the dispersion corrector 130 according to the invention. The addition of the dispersion corrector 130 almost eliminates the nonlinear distortion. The spectrum of FIG. 3D was measured at the output of the DSF 136, and the number and power levels of nonlinear distortion-induced mixing products are significantly reduced compared to the spectrum of FIG. 3C.

The eye diagram in FIG. 4D shows a clear separation between the upper and lower rails. The upper and lower rails correspond to high and low data levels in the demodulated data stream and indicate proper demodulation of the pseudo-random bit sequence. By compensating for the dispersion introduced in the phase conjugator, the present invention provides a significantly better compensation for fiber nonlinearity.

FIG. 5 shows the bit error rate (BER) curve for the system shown in FIG. Hewlett-Packard model as a bit error rate tester for generating curves
No.70842A was used. The baseline BER performance curves corresponding to the spectra and eye diagrams described above in connection with FIGS. 3A and 3B, respectively, have the data points indicated by the x symbols.

The performance for the baseline case shows a receiver sensitivity of about -33 dBm for a bit error rate of 10 ^-9 . The measured BER curves for the signal at the output of the DSF 106 (corresponding to the spectra and eye diagrams of FIGS. 3B and 4B, respectively) have the data points indicated by the square symbols. As mentioned above, the considerable nonlinear distortion introduced by propagating in the 15.4 km DSF 106 at relatively high power levels is
Prevents proper synchronization in the R test equipment.

In order to obtain the BER curve, the total signal power level at the input end of the DSF 106 is 2 dB in increments of +1.
It is lowered to 1 dBm. As a result, the amount of non-linear distortion generated by the DSF 106 is reduced and the BER test equipment can be properly synchronized with the demodulated pseudo-random bit sequence.

The obtained BER curve shows that even at a low input power level, a BER of less than 10 ^-8 has a length of 15.4k.
It also shows that it is not achieved for the demodulated signal propagating in the m DSF 106. About 7 for the baseline case
The dB performance penalty is shown at a BER of about 10 ^-6 .

The BER curve of FIG. 5 with the data points indicated by the O symbols is measured at the output of the system of FIG. 2 using the DSF 122 and the dispersion corrector 130 as the phase conjugator and the dispersion introduced by the DSF 122. To correct. The total signal power level input to the first length DSF 106 is set to +11 dBm. This level is
It is the same power level used to generate the BER curve measured at the output of the DSF 106. 10 performance penalties for baseline cases
^At a BER of ^-9 , it is dramatically reduced to about 1.5 dB.

BER with data points indicated by ▲ symbols
The curve was measured at the output of the system of FIG. 2 with dispersion correction according to the invention and with a total power level of +13 dBm at the input of the DSF 106 of 15.4 km. Therefore, this curve corresponds to the signal spectrum and eye diagram shown in FIGS. 3D and 4D, respectively. Also,
The baseline penalty was only about 1.5 dB.

The results shown in FIGS. 3A-3D, 4A-4D and FIG. 5 demonstrate the performance of systems that use optical phase conjugation to correct for non-linearities and / or chromatic dispersion within an optical fiber span. It shows that the invention can be significantly improved. The results further demonstrate that non-linearity-induced degeneration in WDM systems can be well corrected by using phase conjugation techniques.

For the particular 3-channel WDM system described above, if the center channel is significantly degenerated after transmission with a DSF of 15.4 km, then dispersion-corrected phase conjugation followed by transmission with an additional 14.7 km of DSF Useful for restoring signal integrity.

A 2.5 Gbit / sec modulation center channel already degenerated for the point of suffering a considerable penalty is about the baseline when the dispersion-corrected phase conjugator is used in a fiber span of approximately 30 km. Received with a penalty of only 1.5 dB. Similar effects are obtained with various other optical systems with different span lengths, conversion media and data rates.

The embodiment described with reference to FIG. 2 uses a DSF as the conversion medium for the phase conjugation and a length of standard SMF after the conversion medium for the correction, but this is only an example. However, the present invention is not limited thereto. The correction can be used in phase conjugation and / or frequency shift applications or both or other signal conversion applications. Dispersion correction can be applied at either the input or output end of the conversion medium or distributed throughout the medium. You can also let it.

The non-linear conversion process can be four-photon mixing, three-photon mixing or other non-linear processes.
The conversion medium may be a DSF, SMF or other type of optical fiber, an active semiconductor such as a semiconductor laser or semiconductor optical amplifier, a passive semiconductor, or C.Xu et al., "Efficient bro".
adband wavelength converter for WDM optical commun
ication sysytems, "Technical Digest, OFC '94, pape
r ThQ4, pp. 250-251, San Jose, Cal., a non-linear crystal such as a quasi-phase matching LiNbO ₃ described 1994.

Furthermore, although the main non-linearity causing distortion in the system of FIG. 2 is non-degenerate four-photon mixing of the three-channel signal, the present invention provides self-phase modulation, cross-phase modulation and stimulated Raman scattering (SRS). Similar effects are obtained for removing other non-linearities such as.

FIG. 6 is a graph showing the first-order group velocity dispersion (unit: ps / nm · km) of an optical fiber as a function of wavelength. A large number of first-order group velocity dispersion functions D (λ) are SM
Labeled with F, DSF, DCF1 and DCF2. Each of these first-order group velocity dispersion functions is approximated as a linear function and corresponds to the slope of the linear function shown in FIG.
The second-order variance is approximated as constant.

The dispersion function labeled with the DSF is about 15
It has a zero dispersion at 47 nm and corresponds to the DSF used as the conversion medium in the system of FIG. The function labeled SMF corresponds to the dispersion of standard SMF with zero dispersion at about 1300 nm. As is clear from FIG.
The function DSF has a non-zero dispersion amount at the phase conjugate modulation signal wavelength of about 1539 nm.

The total amount of dispersion introduced by the DSF at the phase conjugate signal wavelength is determined by the length of the DSF used in the phase conjugation. This additional dispersion can be offset, for example, by including an appropriate length of SMF in the signal path of the fiber span after the phase conjugator.

Generally, in a signal conversion using a DSF as a nonlinear medium, when a long frequency input signal is converted into a short frequency signal, a negative dispersion is added to the input end of the conversion medium or a positive dispersion is converted. Must be added to the output end of the medium. The system of Figure 2 corresponds to the latter example.
When a short frequency input signal is converted to a long frequency signal, either positive dispersion must be applied at the input of the conversion medium or negative dispersion must be added at the output of the conversion medium.

FIG. 6 shows the following. That is,
In order to provide a certain amount of positive dispersion at the phase conjugate signal wavelength of 1539 nm, which is suitable for offsetting the negative dispersion introduced by the DSF conversion medium at 1539 nm, an SMF of suitable length at the output of the conversion medium. Can be placed.

In the system of FIG. 2, an SMF of appropriate length is used.
Is determined to be 750 meters. For optical signal converters using other types of non-linear conversion media, the placement of the compensator with respect to the conversion media generally varies depending on the sign of the conversion media and the dispersion introduced by the compensator.

In another embodiment of the invention, the dispersion compensating fiber used exhibits the same zero dispersion as the fiber used as the conversion medium. However, it has a gradient of opposite sign. FIG. 6 is the same as the illustrated DSF function, 1547n.
Shows the dispersion function DCF1 with zero dispersion at m.
Therefore, a fiber of suitable length with the function DCF1 can be placed either before or after the conversion medium DSF in order to correct the dispersion introduced by the medium.

FIG. 6 also shows the dispersion function DCF2. DCF2
Has a large negative dispersion and a large negative slope at 1539 nm. If a DSF is used as the conversion medium, a fiber of suitable length with the function DCF2 is likewise used before the conversion medium for converting short wavelengths, or for long wavelengths, in order to perform the dispersion correction according to the invention. Can be placed after the conversion medium.

In an embodiment of the present invention using an optical fiber nonlinear conversion medium and an optical fiber dispersion compensator, a suitable dispersion correction amount is determined as follows. In this example, the length of the non-linear conversion fiber and the length of the correction fiber are designated L _N and L _C , respectively, and their linear dispersion functions are designated D _N (λ) and D _C (λ), respectively. If the input signal, the pump signal and the phase conjugate signal wavelength are _{denoted by} λ _S , λ _P and λ _S *, respectively, the length L _C of the correction fiber added in front of the nonlinear conversion medium must satisfy the following relationship. . D _N (λ _S ) L _N ≈ −D _C (λ _S ) L _C

The length L _C of the correction fiber added after the nonlinear conversion medium must satisfy the following relationship. D _N (λ _S *) L _N ≈−D _C (λ _S *) L _C Generally, when operating under phase matching conditions, the quantity D
_N (λ _S ) is approximately equal to the quantity −D _N (λ _S *). Furthermore,
Other types of non-linear conversion media, unlike those shown in FIG. 6, have a dispersion function that cannot be linearly approximated linearly, but those skilled in the art can use the technique of the present invention to provide an appropriate dispersion correction amount. It can be easily determined.

FIG. 7 is a block diagram of another embodiment of the present invention. The optical signal converter 200 has an optical signal source 210. The optical signal source 210 provides a single channel or multi-channel input optical signal to the signal combiner 212. The signal combiner 212 receives at least one pump signal from the pump source 214 and combines the input signal and the pump signal into a common signal path. The input signal and pump signal are then provided to the first non-linear element 216.

Nonlinear element 216 provides the partial signal transfer function. The chromatic dispersion introduced into nonlinear element 216 is then partially or completely offset by the opposite amount of dispersion introduced by corrector 218. Similarly, the non-linear elements 220 and 224 provide additional partial signal transfer functions and introduce additional variances that are corrected by the corresponding correctors 222 and 226, respectively. Therefore, the nonlinear element 21
6, 220 and 224 together provide the desired signal transformation such as phase conjugation and / or frequency shift of the input signal. Also, they can be considered as a single optical signal converter such as the phase conjugator 20 of FIG.

The non-linear elements 216, 220 and 224 together correct the dispersion introduced into the signal conversion process. In this embodiment, the dispersion correction is performed by the non-linear element 21.
6, 220 and 224. As in the other embodiments above, the non-linear element is, for example, a dispersion shift of a certain length, a single mode or other type of optical fiber, an active or passive semiconductor, or a non-linear crystal, and the compensator is DS of appropriate length
An F or SMF or any of a number of alternative variance correctors and the like can be used.

In general, correctors 218, 222 and 22
The length of 6 must be chosen so that phase matching is maintained for the pump and signal wavelength used.
However, if the compensators 218, 222, and 226 are constructed from fibers having a dispersion function similar to DCF1 in FIG. 6, phase matching is generally maintained regardless of compensator length.

An alternative to the signal converter 200 of FIG. 7 is:
Nonlinear elements 216, 220 and 224 and corrector 21
8, 222 and 226 can be replaced with one or more lengths of dispersion flattening fiber. Dispersion-flattened fibers exhibit very low chromatic dispersion over the entire suitable wavelength range. In such an embodiment, the dispersion correction is distributed within the conversion medium by selecting a fiber conversion medium that exhibits low negligible dispersion at the desired wavelength. This introduces a minimum variance in the transformed signal.

For the purposes of the present invention, signal transformations involving dispersion-flattening or other types of transformation media specially designed to exhibit low dispersion are considered to be within the scope of the embodiment of FIG. To be disregarded. For a more detailed description of dispersion flattening fibers, see, for example, R. Lundin, "Dispersion Flattening i.
na W Fiber, "Applied Optics, Vol. 33, No. 6, pp.
1011-1014, February 1994.

Another dispersion compensator that can be used in the present invention provides a sufficient amount of chromatic dispersion to partially or completely offset the dispersion produced by the planar dispersion compensator, fiber grating, interferometer or signal conversion. And any other means of introduction. A dispersion compensator suitable for use in the present invention is, for example, C. Poole et al., "Elliptical-Core Dual-M.
ode Fiber Dispersion Compensator, "ECOC '92 Procee
dings, Vol. 3, Post-Deadline Paper No. ThPDI.4, p
p. 863-866, September 1992; K. Takiguchi et al., "D
ispersion Compensation Using a Planar Lightwave Ci
rcuit OpticalEqualizer, "IEEE Photonics Technology
Letters, Vol. 6, No. 4, pp. 561-564; K. Hagimoto
et al., "Penalty free dual-channel 10 Gbit / s trans
missionover 132 km standard fiber using a PLC dela
y equalizer with -830 ps / nm, "OFC '94 Technical Di
gest, Post-Deadline Papers, pp. PD24-1 to PD24-4, S
an Jose, Cal., February 1994; M. Onishi et al., "D
ispersion Compensating Fiber with a Figure of Meri
t of 273 ps / nm / dB and its Compact Packaging, "OFC
'94 Technical Digest, Paper No. 14B1-3, pp. 126-1
27, July 1994; and K. Hill et al., "Aperiodic In-Fi
ber Bragg Gratings for Optical FiberDispersion Com
pensation, "OFC '94 Technical Digest, Post-Deadlin
e Papers, pp. PD2-1 to PD1-4, February 1994, etc.

The optical signal conversion technique of the present invention can be easily extended to a system using two or more signal converters. For example, in such a system, the fiber span can be divided into multiple segments. Each segment has a first and a second portion. An optical phase conjugator is disposed between the first and second portions of each segment to eliminate non-linear effects within the segment.

The non-linearity effect caused by propagating along the first portion of each segment is
Is corrected while propagating along the portion of. For a fiber span divided into n segments, n optical phase conjugators can be used. The length of the first and second parts of each segment and thus the placement of the optical phase conjugator within each segment can be determined in the same manner as for a single segment. In effect, the use of an additional phase conjugator divides the fiber span into shorter length, individually corrected segments.

Therefore, the amount of correction required in each segment is reduced, which improves non-linearity removal. The length of each segment need not be uniform or nearly uniform. For example, a fiber span of length L has a length of 1/3.
It can be divided into two segments, a segment of L and a segment of length 2 / 3L.

The non-linearity removal within each segment is determined by the method described above and the method disclosed in US patent application Ser. No. 08/120014, the relative fiber length, amplifier spacing and signal power in each section. This is done by placing a phase conjugator between the first and second part of each segment having levels.

While the above description illustrates the utility of the present invention in fiber spans incorporating optical phase conjugators to eliminate fiber nonlinearities, the apparatus and method of the present invention generally It is suitable for use with an optical system having an optical phase conjugator that introduces chromatic dispersion into.

For example, the present invention can be used with multi-segment fiber spans incorporating dispersion-corrected optical phase conjugators or other signal converters within each segment of the span. The signal conversion and conversion medium type, the non-linear process used to generate the converted signal, the type of dispersion compensator, the signal conversion application and other system parameters such as data rate and transmission distance can be varied. It can be carried out.

[0098]

As described above, according to the present invention,
The chromatic dispersion introduced as a result of the conversion can be effectively corrected. The present invention corrects chromatic dispersion introduced in signal conversion processes such as optical phase conjugation and / or frequency shift of optical signals. By correcting for this additional chromatic dispersion,
Applications with signal transformations such as midspan optical phase conjugation have the best benefits. Proper compensation of the dispersion introduced by the non-linear medium in systems that use optical phase conjugation to compensate for fiber non-linearities maximizes fiber span bit rate range products. The correction is
For example, dispersion-shifted fiber (DSF), semiconductor laser,
It can be easily and inexpensively provided in a variety of different optical systems using any of a number of nonlinear conversion media such as semiconductor laser amplifiers and nonlinear crystals.

[Brief description of drawings]

FIG. 1 is a block diagram of an example of an optical system that uses midspan optical phase conjugation with dispersion correction according to the present invention.

FIG. 2 is a block diagram of another example of an optical system used to illustrate the improvement provided by using dispersion correction according to the present invention.

3A-3D are waveform diagrams of specific optical spectra generated at various points in the optical system of FIG.

4A-4D are waveform diagrams showing eye diagrams generated at a data rate of 2.5 Gbit / s at various points in the optical system of FIG.

5 is a characteristic diagram of bit error rate (BER) curves showing possible performance improvements in the specific optical system of FIG.

FIG. 6 is a plot showing a specific plot of fiber chromatic dispersion as a function of wavelength in a number of different types of optical fibers.

FIG. 7 is a block diagram of an exemplary embodiment of the invention in which a plurality of dispersion correctors are distributed within a non-linear conversion medium.

[Explanation of symbols]

10 Optical Communication System 12 Optical Transmitter 14 Amplifier 16 Optical Fiber 18 Optical Receiver 20 Optical Phase Conjugate 22 Dispersion Corrector 100 Optical System 101 Transmitter 102, 112, 134 Erbium Doped Fiber Amplifier (EDFA) 104, 114, 126 , 140 band pass filter 108, 116 polarization controller 110 pump 120 signal combiner 124 band reject filter 142 receiver 200 optical signal converter 210 optical signal source 212 signal combiner 214 pump source 216, 220, 224 non-linear element 218, 226 corrector 228 Filter

Claims (36)

[Claims] 1. A device for use in converting an optical signal in an optical signal path, the nonlinear conversion medium being arranged in the optical signal path, capable of receiving the optical signal and generating a converted optical signal therefrom. And at least one dispersion corrector arranged in the optical signal path to provide a suitable amount of chromatic dispersion to offset a portion of the chromatic dispersion introduced into the converted signal by the nonlinear medium,
Optical signal conversion device. 2. The optical signal path has an optical fiber span, the nonlinear conversion medium is part of an optical phase conjugator arranged in the span, and the converted optical signal is a phase conjugate of the optical signal. Equipment. 3. The optical signal path has an optical fiber span, the non-linear conversion medium is part of a frequency shifter arranged in the span, and the converted optical signal is a frequency shifted version of the optical signal. The apparatus according to item 1. 4. A pump signal source for providing a pump signal and a combination of the optical signal and the pump signal so as to mix the optical signal and the pump signal in a non-linear conversion medium to produce a mixing product containing the converted optical signal. The apparatus of claim 1 further comprising a signal combiner arranged in the optical signal path before the non-linear conversion medium to effectuate. 5. The apparatus of claim 1, wherein the non-linear conversion medium is a length of dispersion-shifting fiber. 6. A length of dispersion-shifting fiber is about 0.
The apparatus of claim 5 having a second-order group velocity dispersion in the range of 04 to 0.10 ps / nm ² · km. 7. The device of claim 1, wherein the non-linear conversion medium is an active or passive semiconductor. 8. The apparatus of claim 1, wherein the non-linear conversion medium is a non-linear crystal. 9. The apparatus of claim 1, wherein the dispersion corrector is located in the optical signal path before the nonlinear conversion medium. 10. The apparatus of claim 1, wherein the dispersion corrector is located in the optical signal path after the nonlinear conversion medium. 11. The apparatus of claim 1, wherein the dispersion corrector is distributed within the non-linear conversion medium. 12. The optical signal is a multi-channel optical signal and the dispersion corrector is introduced by a non-linear conversion medium over a wavelength range that includes the wavelength of the channel signal within the multi-channel optical signal or the wavelength of the phase conjugation of the channel signal. The apparatus of claim 1, providing a chromatic dispersion having a sign opposite to that of the chromatic dispersion. 13. A step of inputting an optical signal into a non-linear conversion medium arranged in an optical signal path of an optical signal, a step of generating a converted optical signal in the non-linear conversion medium, and introducing into the converted signal by the non-linear conversion medium. The method of converting an optical signal, comprising the step of correcting the chromatic dispersion introduced into the nonlinear conversion medium by supplying an appropriate chromatic dispersion amount for offsetting a part of the chromatic dispersion generated. 14. The method further comprising the steps of providing a pump signal and mixing the optical signal and the pump signal in a non-linear conversion medium to generate a converted optical signal.
Method 3 15. The step of inputting an optical signal comprises inputting the optical signal into a non-linear conversion medium in an optical phase conjugator arranged in an optical fiber span, the step of generating the converted optical signal comprising: 14. The method of claim 13 including generating a phase conjugate of the optical signal. 16. The step of inputting an optical signal comprises inputting the optical signal into a non-linear conversion medium within a frequency shifter arranged in an optical fiber span, the step of generating the converted optical signal comprising: The method of claim 13, comprising generating a frequency-shifted version of the optical signal. 17. The method of claim 13 wherein the step of inputting the optical signal into the nonlinear conversion medium comprises inputting the optical signal into a length of dispersion-shifting fiber. 18. The method of claim 13, wherein the step of inputting the optical signal to the nonlinear conversion medium comprises inputting the optical signal to an active or passive semiconductor. 19. The method of claim 13 wherein the step of inputting the optical signal into the nonlinear conversion medium comprises inputting the optical signal into a nonlinear crystal. 20. The step of correcting the chromatic dispersion introduced by the non-linear conversion medium comprises providing a dispersion corrector in the optical signal path of the optical signal before the non-linear conversion medium.
Method 3 21. The step of correcting chromatic dispersion introduced by a non-linear conversion medium comprises providing a dispersion corrector in the optical signal path of the optical signal after the non-linear conversion medium.
Method 3 22. The method of claim 13, wherein the step of correcting the chromatic dispersion introduced by the non-linear conversion medium comprises providing a dispersion compensator distributed within the non-linear conversion medium. 23. Providing an optical fiber span comprised of at least one segment, each segment having a first portion and a second portion, for providing an optical signal to the optical fiber span, Providing an optical signal transmitter at one end of the optical fiber span,
Providing an optical signal receiver at the other end of the optical fiber span for receiving an optical signal from the optical fiber span, and providing an optical phase conjugation provided between the first and second portions of the segment Phase conjugating the optical signal in the chamber, selecting the power level of the optical signal in at least one portion of at least one segment of the fiber span to correct for the nonlinearity of the fiber span, and An optical signal transmission method comprising the step of correcting the amount of chromatic dispersion generated by the phase conjugation step. 24. The step of correcting the amount of chromatic dispersion comprises an additional length of optical fiber having an amount of chromatic dispersion suitable for offsetting at least a portion of the chromatic dispersion introduced into the phase conjugate optical signal by the phase conjugator. 3. Supplying
Method 3 25. The method of claim 23, wherein the step of selecting a power level sufficient to correct for span non-linearity comprises adjusting the output power of the in-line amplifier in at least one portion. 26. An optical fiber span comprised of at least one segment, each segment having a first portion and a second portion, at one end of the optical fiber span for supplying an optical signal to the optical fiber span. An optical signal transmitter, the optical signal having a power level selected such that the non-linearity of the fiber span is corrected within said portion, the other end of the optical fiber span for receiving the optical signal from the optical fiber span , An optical phase conjugator arranged between the first and second parts of the segment for phase conjugating the optical signal, and an amount of chromatic dispersion introduced by the optical phase conjugator An optical communication system comprising a dispersion compensator arranged in a segment for compensating for. 27. The system of claim 26, wherein the optical phase conjugator comprises a length of dispersion-shifting fiber and the dispersion corrector is a length of single-mode fiber. 28. The system of claim 26, wherein the power level of the optical signal is selected by adjusting the output power of the in-line amplifier in at least one portion of the plurality of portions. 29. A non-linear conversion medium arranged in the optical signal path and capable of receiving the optical signal, a pump signal source for supplying the pump signal to the non-linear conversion medium, whereby the pump signal and the optical signal are in the non-linear conversion medium. A device for converting an optical signal, comprising: mixing, generating a converted optical signal, a non-linear conversion medium selected to introduce a minimum chromatic dispersion into the converted optical signal. 30. The apparatus of claim 29, wherein the non-linear conversion medium includes dispersion corrections distributed within the medium. 31. The non-linear conversion medium comprises a plurality of dispersion-shifting fibers of length and at least one dispersion corrector arranged in an optical signal path between the dispersion-shifting fibers of a plurality of lengths. Equipment. 32. The apparatus of claim 29, wherein the non-linear conversion medium is a length of dispersion-flattening fiber. 33. Providing a non-linear conversion medium arranged in the optical signal path and capable of receiving the optical signal; providing a pump signal to the non-linear conversion medium; and generating a converted optical signal. A method of converting an optical signal comprising the steps of mixing an optical signal and a pump signal in a non-linear conversion medium and selecting the non-linear conversion medium such that a minimum chromatic dispersion is introduced into the converted optical signal. 34. The method of claim 33, wherein the step of selecting a non-linear conversion medium comprises selecting a medium having dispersion corrections distributed within the medium. 35. The step of selecting a non-linear conversion medium comprises a medium having dispersion-shifting fibers of multiple lengths and at least one dispersion correction arranged in the optical signal path between the dispersion-shifting fibers of multiple lengths. 35. Selecting a vessel
the method of. 36. The method of claim 33, wherein the step of selecting a non-linear conversion medium comprises selecting a medium consisting of a length of dispersion-flattened fiber.